Date post: | 05-Feb-2017 |
Category: |
Documents |
Upload: | truongnhan |
View: | 228 times |
Download: | 2 times |
1
15.053/8 April 11, 2013
Introduction to Networks
2
Quotes for today
"A journey of a thousand miles begins with a single step." -- Confucius “You cannot travel the path until you have become the path itself” -- Buddha
3
Network Models
Optimization models
Can be solved much faster than other LPs
Applications to industrial logistics, supply chain management, and a variety of systems
Today’s lecture: introductory material, Eulerian tours, the Shortest Path Problem
Application of Network Models: http://jorlin.scripts.mit.edu/docs/publications/52-
applications%20of%20network.pdf
4
Notation and Terminology Note: Network terminology is not (and never will be) standardized. The same concept may be denoted in many different ways.
Called: • NETWORK • directed graph • digraph • graph
2
1
4
3
2
1
4
3
Class Handouts (Ahuja, Magnanti, Orlin)
Node set N = {1, 2, 3, 4} Arc Set
Network G = (N, A)
{(1,2), (1,3), (3,2), (3,4), (2,4)}
Graph G = (V, E)
Edge set E
Also Seen
Vertex set V
5
Directed and Undirected Networks
2
3 4
1
a
b
c
d
e
An Undirected Graph
2
3 4
1
a
b
c
d
e
A Directed Graph
The field of Network Optimization concerns optimization problems on networks
Networks are used to transport commodities • physical goods (products, liquids) • communication • electricity, etc.
6
Networks are Everywhere Physical Networks
– Road Networks – Railway Networks – Airline traffic Networks – Electrical networks, e.g., the power grid – Communication networks
Social networks
– Organizational charts – friendship networks – interaction networks (e.g., cell calls)
7
Overview:
Most O.R. models have networks or graphs as a major aspect
Next two lectures: focus on network optimization problems.
Next: representations of networks
– Pictorial
– Computer representations
LOST: An
Illustrative
Example
Image removed due to copyright restrictions.
See interactive graphic “The Web of Intrigue” in “As Lost Ends, Creators Explain How They Did It, What’s Going On.” Wired Magazine, April 19, 2010.
Wired has a handy character chart (yes, there might be spoilers!), created by bioinformics scientist Martin Krzywinski using Circos software (even that sentence is confusing) that shows how all of the characters are related. For example, if you're wondering how many of the characters are related via "romance," click on Romance and the chart will change to show you that. Same with Chance, Family, Occupational, Touched By Jacob, Undisclosed.
An elegant representation of arcs
http://flare.prefuse.org/apps/dependency_graph
© UC Berkeley Visualization Lab. All rights reserved. This content is excluded from our CreativeCommons license. For more information, see http://ocw.mit.edu/help/faq-fair-use/.
Graph of Ph.D. advisors
James Orlin (1981)
Arthur F. Veinott, Jr. (1960)
Cyrus Derman (1954)
Herbert E. Robbins (1938)
Hassler Whitney (1932)
George Birkoff (1907)
E.H. Moore (1885)
H.A Newton (1850)
Michel Chasles (1814)
Simeon Denis Poisson (1800)
Joseph Louis Lagrange (1754)
Leonhard Euler (1726)
Johann Bernoulli (1694)
Jacob Bernoulli (1676)
Nicolas Malebranche (1672)
Gottfried Leibniz (1666)
11
The Adjacency Matrix (for directed graphs)
2
3 4
1 a
b
c
d
e
A Directed Graph
•Have a row for each node
1 2 3 4 1 2 3 4
•Have a column for each node •Put a 1 in row i- column j if (i, j) is an arc What would happen if (4, 2) became (2, 4)?
12
The Adjacency Matrix (for undirected graphs)
•Have a row for each node •Have a column for each node •Put a 1 in row i- column j if (i, j) is an arc
2
3 4
1 a
b
c
d
e
An Undirected Graph
The degree of a node is the number of incident arcs
degree 2 3 2 3
1 2 3 4 1 2 3 4
Note: each arc shows up twice in the adjacency matrix.
✓
Question. Is it possible that the number of nodes of odd degree is odd?
13
1. Yes
2. No
14
Arc list representations
2
3 4
1 a
b
c
d
e
A Directed Graph
1: (1,2), (1,4) 2: (2,3) 3: ∅ 4: (4,2), (4,3)
(1,2) (1,4) (2,3) (4,2) (4,3)
1 2 3 4 Nodes
Arcs
Forward Star Representation.
Node i points to first arc on arc list whose head is node i.
Which uses computer space more efficiently for large road networks: the adjacency matrix or adjacency lists?
15
✓ 1. Adjacency matrix
2. Adjacency lists
e.g. consider a road network with 10,000 nodes, and with 40,000 arcs
The adjacency matrix has 100 million entries.
The adjacency list has at most 80,000 entries, two for each road.
16
On network representations
Each representation has its advantages – Major purpose of a representation
• efficiency in algorithms • ease of use
Next: definitions for networks
17
Directed Path . Example: 1, 2, 5, 3, 4 (or 1, a, 2, c, 5, d, 3, e, 4)
•No node is repeated. •Directions are important.
Cycle (or circuit or loop) 1, 2, 3, 1. (or 1, a, 2, b, 3, e)
•A path with 2 or more nodes, except that the first node is the last node. •Directions are ignored.
Path: Example: 5, 2, 3, 4. (or 5, c, 2, b, 3, e, 4)
•No node is repeated. •Directions are ignored.
Directed Cycle: (1, 2, 3, 4, 1) or 1, a, 2, b, 3, c, 4, d, 1
•No node is repeated, except that the first node is the last node. •Directions are important.
2 3 4 a b c
1
5 d e
2
3
4
a b
c d 1 e
2 3 4 a b c
1
5 d e
2
3
4
a b
c d 1 e
18
Walks
2
3 4
1
a
b
c
d
e
5
2
3 4
1
a
b
c
d
e
5
Walks are paths that can repeat nodes and arcs Example of a directed walk: 1-2-3-5-4-2-3-5 A walk is closed if its first and last nodes are the same. A closed walk is a cycle except that it can repeat nodes and arcs.
19
More terminology
An undirected network is connected
if every node can be reached from every other node by a path
2
1
4
3
5
2
1
4
3
5
A directed network is connected if it’s undirected version is connected.
This directed graph is connected, even though there is no directed path between 2 and 5.
20
On connectivity
6
1
4
3
7
2
10
8
9
5
There are simple efficient procedures for determining if a graph is connected.
Here is a graph with two components, that is maximally connected subgraphs.
4 7
10 9
We will not describe these algorithms, but will do a more general algorithm later in this lecture
21
The Bridges of Koenigsberg: Euler 1736
“Graph Theory” began in 1736 Leonard Euler
– Visited Koenigsberg – People wondered whether it is
possible to take a walk, end up where you started from, and cross each bridge in Koenigsberg exactly once
– Generally it was believed to be impossible
22
The town of Koenigsberg
A
B
D
C
Annotated map © source unknown. All rights reserved. This content is excluded from ourCreative Commons license. For more information, see http://ocw.mit.edu/help/faq-fair-use/.
23
The Bridges of Koenigsberg: Euler 1736
A
D
C B
1 2
4
3
7
6 5
Is it possible to start in A, cross over each bridge exactly once, and end up back in A?
24
The Bridges of Koenigsberg: Euler 1736
A
D
C B
1 2
4
3
7
6 5
Conceptualization: Land masses are nodes
25
The Bridges of Koenigsberg: Euler 1736
1 2
4
3
7
6 5
Conceptualization: Bridges are arcs
A
C
D
B
26
The Bridges of Koenigsberg: Euler 1736
1 2
4
3
7
6 5
Translation to graphs or networks: Is there a walk starting at A and ending at A and passing through each arc exactly once? Why isn’t there such a walk?
A
C
D
B
27
Adding two bridges creates such a walk
A, 1, B, 5, D, 6, B, 4, C, 8, A, 3, C, 7, D, 9, B, 2, A
1 2 4
3
7
6 5
A
C
D
B
8
9
Here is the walk.
Note: the number of arcs incident to B is twice the number of times that B appears on the walk.
28
Eulerian cycle: a closed walk that passes through each arc exactly once
Degree of a node = number of arcs incident to the node
Necessary condition: each node has an even degree.
Why necessary? The degree of a node j is twice the number of times j appears on the walk (except for the initial and final node of the walk.)
Theorem. A graph has an eulerian cycle if and only if
the graph is connected and every node has even
degree.
29
Eulerian path: a walk that is not closed and passes through each arc exactly once
Theorem. A graph has an Eulerian path if and only if exactly two nodes have odd degree and the graph is connected.
30
Eulerian cycles
Eulerian cycles and extensions are used in practice
Mail Carrier routes: – visit each city block at least once – minimize travel time – other constraints in practice?
Trash pickup routes – visit each city block at least once – minimize travel time – other constraints in practice?
Traveling Salesman Problem The 48 city problem.
31 George Dantzig, Ray Fulkerson, and Selmer Johnson (1954)
Mental Break
32
33
More Definitions
A network is connected if every node can be reached from every other node by a path
2
1
4
3
5
A spanning tree is a connected subset of a network including all nodes, but containing no cycles.
2
1
4
3
5
2
1
4
3
5
2
1
4
3
5
34
More on Trees An out-tree is a spanning tree in which every node has exactly
one incoming arc except for the root.
Theorem. In an out-tree, there is a directed path from the root to all other nodes. (All paths come out of the root).
One can find the path by starting at the end and working backwards.
2
1
4
3
5
2
1
4
3
5
7 8 9
6
10 11
12 13 13
35
The Shortest Path Problem
1
2
3
4
5
6
2
4
2 1
3
4
2
3
2
What is the shortest path from a source node (often denoted as s) to a sink node, (often denoted as t)? What is the shortest path from node 1 to node 6? Assumptions for this lecture:
1. There is a path from the source to all other nodes. 2. All arc lengths are non-negative
36
Shortest Path Problem
Where does it arise in practice? – Common applications
• shortest paths in a vehicle (Navigator) • shortest paths in internet routing • shortest paths around MIT
– and less obvious applications, as in the course readings (see URL on slide 3 of this lecture).
How will we solve the shortest path problem? – Dijkstra’s algorithm
Application 1: Shortest paths in a Transportation Network
37
Add a node for every “intersection”. Add arcs for roads.
38
Dijkstra’s Algorithm
Exercise: find the shortest path from node 1 to all other nodes. Keep track of distances using labels, d(i) and each node’s immediate predecessor, pred(i).
d(1)= 0, pred(1)=0;
d(2) = 2, pred(2)=1
Find the other distances, in order of increasing distance from node 1.
1
2
3
4
5
6
2
4
2 1
3
4
2
3
2
Exercise: Find the shortest paths by inspection.
39
Key observations Suppose that d(i) is the length of some path from node 1
to node i. Suppose that there is an arc (i, j) of length cij. Then there is a path from node 1 to node j of length at
most d(i) + cij.
10
In this case, there is a path from 1 to j of length 72. We can reduce d(j) to 72.
i j d(i) = 62
1 P
Length(P) = 62
P’ Length(P’) = 78 d(j) = 78
40
A Key Procedure in Shortest Path Algorithms
At each iteration d(j) is the length of some path from node 1 to node j. (If no path is known, then d(j) = ∅)
78
Up to this point, the best path from 1 to j had length 78 But P, (i,j) is a path from 1 to j of length 72.
72
Procedure Update(i) for each (i,j) ∈ A(i) do if d(j) > d(i) + cij then d(j) : = d(i) + cij and pred(j) : = i;
i j 62 10
1 P
41
Dijkstra’s Algorithm
begin d(s) : = 0 and pred(s) : = 0; d(j) : = ∅ for each j ∈ N - {s}; LIST : = {s}; while LIST ≠ ∅ do begin let d(i) : = min {d(j) : j ∈ LIST}; remove node i from LIST; update(i) if d(j) decreases from ∞,
place j inLIST end end
Initialize distances.
LIST = set of temporary nodes
Select the node i on LIST with minimum distance label, and then update(i)
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(2) = ∞ pred(2) = ∅
LIST = {1,
d(1) = 0 pred(1) = ∅
d(4) = ∞ pred(4) = ∅
d(3) = ∞ pred(3) = ∅
d(5) = ∞ pred(5) = ∅
d(6) = ∞ pred(6) = ∅
Initialize
d( ) = {0,
43
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(2) = ∞ pred(2) = ∅
LIST = {1,
d(1) = 0 pred(1) = ∅
d(4) = ∞ pred(4) = ∅
d(3) = ∞ pred(3) = ∅
d(5) = ∞ pred(5) = ∅
d(6) = ∞ pred(6) = ∅
Find the node i on LIST with minimum distance label.
d( ) = {0,
Remove i from LIST. Make i permanent.
LIST = {
d( ) = {
1
44
d(3) = ∞ pred(3) = ∅
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(2) = ∞ pred(2) = ∅
LIST = {1,
d(1) = 0 pred(1) = ∅
d(4) = ∞ pred(4) = ∅
d(3) = 4 pred(3) = 1
d(5) = ∞ pred(5) = ∅
d(6) = ∞ pred(6) = ∅
update(1)
d( ) = {0,
LIST = {
d( ) = {
1
d(2) = 2 pred(2) = 1
LIST = { 2,
d( ) = { 2
LIST = { 2, 3
d( ) = { 2, 4
1
45
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(2) = pred(2) =
LIST = {1,
d(1) = 0 pred(1) = ∅
d(4) = ∞ pred(4) = ∅
d(3) = 4 pred(3) = 1
d(5) = ∞ pred(5) = ∅
d(6) = ∞ pred(6) = ∅
d( ) = {0,
LIST = {
d( ) = {
1
d(2) = 2 pred(2) = 1
LIST = { 2,
d( ) = { 2
LIST = { 2, 3
d( ) = { 2, 4
Find the node i on LIST with minimum distance label.
Remove i from LIST. Make i permanent.
LIST = { 3
d( ) = { 4
2
Arcs (2, 3), (2, 4) and (2,5) will be scanned next. Which nodes will have their distance label changed?
46
1
2
3
4
5
6
2
2 1
3 4
2
3 2
d(2) = 2
d(3) = 4 d(5) = ∞
d(6) = ∞ 1
2 d(4) = ∞
d(1) = 0
4
✓ 1. 2, 3, 4 and 5
2. 3, 4, and 5
3. 4 and 5
4. none of the above.
47
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
d(4) = ∞ pred(4) = ∅
d(3) = 4 pred(3) = 1
d(5) = ∞ pred(5) = ∅
d(6) = ∞ pred(6) = ∅
1
Update(2)
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
2
LIST = { 3,
d( ) = { 4
LIST = { 3,
d( ) = { 3
LIST = { 3, 4
d( ) = { 3, 6
LIST = { 3, 4, 5
d( ) = { 3, 6, 4
48
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
d(6) = ∞ pred(6) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
LIST = { 3, 4, 5
d( ) = { 3, 6, 4
d(2) = 2 pred(2) = 1
Find the node i on LIST with minimum distance label.
Remove i from LIST. Make i permanent.
3
LIST = {4, 5
d( ) = {6, 4
49
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
d(6) = ∞ pred(6) =∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
LIST = {4, 5
d( ) = {6, 4
3
Update(3)
50
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
d(6) = ∞ pred(6) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
LIST = {4, 5
d( ) = {6, 4
Find the node i on LIST with minimum distance label.
Remove i from LIST. Make i permanent.
5
LIST = {4
d( ) = {6
51
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
d(6) = ∞ pred(6) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
LIST = {4
d( ) = {6
5
Update(5)
d(6) = 6 pred(6) = 5
LIST = {4, 6
d( ) = {6, 6
52
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
1
2
d(4) = 6
5
d(6) = 6
Which node will be scanned next according to the usual rule?
✓
1. node 4
2. node 6
3. either node 4 or node 6; both choices are OK.
53
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {4, 6
d( ) = {6, 6
Find the node i on LIST with minimum distance label.
Remove i from LIST. Make i permanent.
LIST = {6
d( ) = {6
4
54
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {6
d( ) = {6
4
Update(4)
55
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {6
d( ) = {6
Find the node i on LIST with minimum distance label.
Remove i from LIST. Make i permanent.
LIST = {
d( ) = {
6
56
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {6
d( ) = {6
LIST = {
d( ) = {
6
Update(6)
Node 6 has no outgoing arcs.
57
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {6
d( ) = {6
LIST = {
d( ) = {
Find the node i on LIST with minimum distance label.
LIST = ∅. The algorithm ends.
58
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
The shortest path from node 1 to node 6.
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {6
d( ) = {6
LIST = {
d( ) = {
Trace back the path from node 6 to node 1 using the predecessors.
59
1
2
3
4
5
6
2
4
2 1
3
4
2
3 2
d(1) = 0 pred(1) = ∅
The shortest path from node 1 to node 6.
1
2
d(3) = 3 pred(3) = 2
d(4) = 6 pred(4) = 2
d(5) = 4 pred(5) = 2
d(2) = 2 pred(2) = 1
d(6) = 6 pred(6) = 5
LIST = {6
d( ) = {6
LIST = {
d( ) = {
The “predecessor” arcs form an out-tree rooted at node 1.
60
Comments on Dijkstra’s Algorithm Dijkstra’s algorithm makes nodes permanent in
increasing order of distance from the origin node.
Dijkstra’s algorithm is efficient in its current form. The running time grows as n2, where n is the number of nodes
It can be made much more efficient
In practice it runs in time linear in the number of arcs (or almost so).
61
Edsger Dijkstra 1930-2002
Turing Prize 1972 • development of Algol
• programming languages
• graph theory
http://en.wikipedia.org/wiki/Edsger_W._Dijkstra
62
Summary
The Eulerian cycle problem
The shortest path problem
Dijkstra’s algorithm finds the shortest path from node 1 to all other nodes in increasing order of distance from the source node.
The bottleneck operation is identifying the minimum distance label. One can speed this up, and get an incredibly efficient algorithm
MIT OpenCourseWarehttp://ocw.mit.edu
15.053 Optimization Methods in Management ScienceSpring 2013
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.